Automated Algolia Index Health Monitoring
Every morning, WebRun opens the Algolia dashboard, reads each index's record count, last build time and operation volume, compares them against yesterday to catch indexes that shrank sharply or stopped updating, posts a health table to Microsoft Teams, and queues a Twilio text to the on-call engineer when an index looks genuinely broken.
How do I know if my Algolia search index stopped updating?
WebRun checks every Algolia index each morning, reading its record count, last update time and search volume, then comparing them against yesterday. It posts a health table to Microsoft Teams with failing indexes first, and queues a Twilio text to on-call when an index has shrunk sharply or stopped updating.
- A broken indexing job is caught the next morning, not by a customer
- Record count drops are compared against yesterday, not guessed at
- On-call is only texted when an index genuinely broke
Built for engineering teams · site reliability engineers · ecommerce developers · platform teams
What does WebRun do on every run?
The exact actions WebRun takes, in order - in plain language, so you can adjust anything.
-
WebRun signs in and gets to work
Opens
www.algolia.com/users/sign_inin a real browser with your saved login - no setup, no API keys. -
1
Algolia - read index counts and build times
WebRun opens Algolia to read index counts and build times. - Open the Algolia dashboard and list every index in the application
- Capture the record count, index size and the time of the last update for each
- Compare today's record count against yesterday's and flag drops beyond the percentage you set
- Flag any index with no update in longer than its normal refresh cycle
- Read the usage view for search operations and record any index whose query volume fell to near zero
- Note replica indexes separately so an expected mirror is not read as a problem
Done when Every index has a record count, a last update time and a verdict against yesterday's numbers.
-
2
Microsoft Teams - post the health table
WebRun opens Microsoft Teams to post the health table. - Post a health table to your engineering channel with one row per index
- Show record count, change against yesterday, last update time and search volume
- Put failing indexes at the top with the reason spelled out
- Say plainly when every index passed, so a green day is still recorded
Done when Today's Algolia health table is in Microsoft Teams.
-
3
Twilio - text on-call when an index breaks
WebRun opens Twilio to text on-call when an index breaks. - When an index drops records past your threshold or has gone stale, draft a short text to the on-call engineer
- Name the index, the record count before and after, and the last update time
- Queue the text for approval before it sends, since it wakes a real person on a real phone
- Never delete, rebuild, clear or reconfigure an index. WebRun reads the dashboard and reports
Done when An alert text is queued for on-call, or no index broke and nothing was queued.
How is each run configured?
Secure by default
Connect once, stays signed in
WebRun signs in once and keeps each session in a persistent environment, so every run picks up right where it left off.
Every action is checked against this policy before it runs.
Questions, answered
Will it rebuild or clear an index?
No. WebRun reads the Algolia dashboard only. Clearing, rebuilding, deleting and reconfiguring an index all change live search results, so they stay with your engineers. The run reports and alerts.
Will it text on-call without approval?
No. The Twilio alert is queued for approval before it sends, because it reaches a real phone. Internal posts to Microsoft Teams go up on their own, since those are just a channel message.
What counts as a broken index?
A record count that fell past the percentage threshold you set, or an index with no update in longer than its normal refresh cycle. Replica indexes are tracked separately so an expected mirror is not flagged.
Put this on autopilot.
Turn it on in minutes - or have our team set it up for you.